On The Mining and Usage of Movement Patterns in Large Traffic Networks

Al-Zeyadi, Mohammed, Coenen, Frans ORCID: 0000-0003-1026-6649 and Lisitsa, Alexei
(2017) On The Mining and Usage of Movement Patterns in Large Traffic Networks. In: 2017 IEEE International Conference on Big Data and Smart Computing (BigComp), 2017-2-13 - 2017-2-16.

[img] Text
bigcompIEEE_2016-10-17.pdf - Author Accepted Manuscript

Download (995kB)


Paper presents the Shape Movement Pattern (ShaMP) algorithm, an algorithm for extracting Movement Patterns (MPs) from network data, and a prediction mechanism whereby the identified MPs can be used to predict the nature of movement in a previously unseen network. The principal advantage offered by ShaMP is that it lends itself to parallelisation. The reported evaluation was conducted using both Massage Pass Interface (MPI) and Hadoop/MapReduce; and artificially generated and real life networks. The later extracted from the UK Cattle tracking Systems (CTS) in operation in Great Britain (GB). The evaluation indicates that very successful results can be produced, average precision, recall and F1 values of 0.965, 0.919 and 0.941 were recorded respectively.

Item Type: Conference or Workshop Item (Unspecified)
Uncontrolled Keywords: Big Network Data, Pattern Mining, Movement Patterns, Prediction, Hadoop, MPJ Express
Depositing User: Symplectic Admin
Date Deposited: 08 Feb 2017 08:54
Last Modified: 19 Jan 2023 07:19
DOI: 10.1109/bigcomp.2017.7881729
Related URLs:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3005604